By Stéphane Gallot, Pascal Dutot, and Christophe Sajous
A new hardware assessment tool automates testing and mission replay, managing military GPS receiver input and output data, with an operational implementation and with a better control of initialization conditions, especially direct P(Y) acquisition. The test bench drives a GPS/Galileo simulator, a digital jammer, and software programs for visibility computation based on terrain modeling, and for multipath generation on 3D renderings.
Comprehensive assessment of military GPS receivers becomes more complex as they are integrated into advanced systems. To limit testing on systems under live conditions, laboratory evaluations with real elements are essential.
A new hybrid test bench called Statistical INERtial Gnss HYbrid in Simulation (SINERGHYS) is designed for governmental use to validate the integration of GPS/Galileo receivers within the navigation system for different platforms. As system-level requirements become more stringent, this bench has been designed to assess the behavior of the complete system in an operational context.
This new assessment hardware-in-the-loop tool is designed to automate testing and to replay missions with an operational implementation and with a better control of initialization conditions, especially direct P(Y) acquisition. This test bench drives many simulation tools: a GPS/Galileo simulator, a digital miniaturized jammer, and different softwares such as one enabling the computation of visibility depending on the terrain modeling, or one dedicated to the generation of multipaths on surfaces of realistic 3D scenes.
A Common Bench. Since 2000, with the arrival of the new cryptographic generation (the selective availability anti-spoofing module, or SAASM), the French government defence procurement agency (DGA) GPS laboratory decided to buy off-the-shelf GPS SAASM receivers that cover different form factors and applications. To test performance, it was necessary to acquire a test bench suitable for each GPS receiver. Testing procedures became more and more complex, and most of the manufacturer-provided benches could not perform every test required, such as direct P(Y) acquisition. To improve French expertise concerning GPS receivers, the DGA GPS laboratory decided to develop a common, generic test bench taking into account the integration constraints of each receiver. The perimeter of the hybrid test bench consists of a PC and a generic GPS test bench.
Figures 3 and 4 show examples of military GPS receivers integrated into the bench.
Bench management is centralized, so test conditions are generic, and all simulation parameters are fully controlled. This enables users to display a unique view of the complete information and to be able to replay specific scenarios.
The bench manages military GPS receivers’ input and output data as described in the respective receivers’ interface control document (ICD) or interface specification: this enables, for example, the initialization of GPS receivers by sending precise time to facilitate direct P(Y) acquisition. This new bench is compatible with many GPS receivers with different form factors and applications.
Several receivers can be tested at the same time with the same software, so that the behavior of the GPS receivers can be compared in real time. Data from the different receivers can be observed on the same window of the graphic user interface (GUI). Specific data from ICDs can be displayed on the GUI. The user can visualize three different windows: the first is related to integrity, the second to alarms, and the third to cryptography. All the data output by the receivers can be recorded and replayed.
To facilitate and enhance trials on GPS receivers, the bench can use a Monte Carlo method, enabling sequentially and automatically chained scenarios, up to 10,000 test sequences, primarily for characterization of time-to-first-fix (TTFF).
Inertial navigation system (INS)/GPS hybridization in real time can be simulated via processing based on a Kalman filter of the information delivered by simulated INS and GPS. Loose and tight coupling can be selected through the GUI as well as filter parameters. The Kalman filter design is independent from the receiver and from the type of trajectory simulated. The user can decide whether the GPS receiver does receive aiding either from the simulated INS, or from the optimal navigation (output of Kalman filter).
The bench can interface with various external means and drive some tools and materials involved in the functioning of the bench.
With GPS Simulator. In the interface with the simulator, an intuitive GUI facilitates scenario preparation. When ready, SINERGHYS begins to drive the GPS simulator in remote-control mode. Any type of trajectory can be simulated with its operational environment modeled. The simulator outputs an RF signal to the receiver, and representative aiding, if required, by ethernet protocol to SINERGHYS.
With Jammer. Two types of interference signal generators can be used with the bench. Any available waveform can be generated. The bandwidth can go up to 20 Mhz for one generator and up to 80 Mhz for the other.
SINERGHYS is also compatible with a specific jammer called Embedded Jammer, designed to test vulnerability of GNSS systems (Figure 5).
The GPS receiver under test tracks the real GPS satellites combined with the simulated jamming signal. Thanks to the position and attitudes provided by the aircraft and to a modelized antenna diagram, the jammer computes in real time representative jamming that would be generated by real jammers.
This jammer works in two modes: localized mode (coordinates, jammer power, and waveform) and power profile mode. It was initially designed to be used inside an aircraft but can be used for laboratory testing as well.
The simulated environment is defined in the configuration software: waveform, emitter, scenario definitions (bands, number of emitters), and antenna diagram.
Four GNSS bands can be selected: GPS L1 and L2 (40 MHz) and Galileo E6 (40 MHz) and E5 (90 MHz). The embedded jammer can generate up to 14 simultaneous jammers per band, each with different waveforms. Therefore, up to 56 simultaneous jammers can be simulated.
The center frequency of the jamming signals can be chosen anywhere in the bandwidth. Modulation examples: continuous wave, broadband noise, binary phase shift keying), binary offset carrier (x,y), and so on.
External software interfaces fall under three categories.
Warfare. Electronic warfare software, which provides jamming coverage, performs a precise assessment of propagation (reflection and diffraction) of the interfering signals (depending on terrain modeling). Interference levels are transmitted to SINERGHYS during pre-processing.
Satellite Tool Kit (STK). This software is designed to provide sophisticated modeling and visualization capabilities and performs functions critical to all mission types, including propagation of vehicles, and determination of visibility areas and times. STK generates paths for space and ground-based objects, such as satellites, ships, aircraft, and land vehicles. STK also provides animation capabilities and a two-dimensional map background for visualizing the path of these vehicles. Within SINERGHYS, STK is used for real-time visualization.
Ergospace. This software is designed to generate multipaths, enabling the modeling of reflected paths of different satellite signals on surfaces of realistic 3D scenes. Pre-processed multipaths are sent to SINERGHYS and generated by the GPS simulator. The software is also used for real-time visualization.
Operational Mission Characterization
The bench can evaluate and characterize receiver performance in most possible representative conditions.
Management of GPS Inputs/Outputs. Both black and red keys can be loaded inside the GPS receivers in both DS101 and DS102 protocols. This loading can be performed manually through key loaders such as KYK13 or DTD/ANCYZ10, but also through the host application with hexadecimal keys.
The bench can send commands to GPS receivers such as non-volatile memory erasure command, INS, precise time source, precise time and time interval (PTTI) activation commands, or choices between “mixed mode” and “all Y,” between “L1 primary” and “L2 primary,” and so on. Depending on user requirements, the bench can provide time, position, speed, almanac, ephemeris, or specific navigation sub-frames.
To test the jamming resistance of GPS receivers, it is essential to be able to provide INS aiding. SINERGHYS uses perfect or degraded aiding and adapts the format or the frequency for the considered GPS receiver.
Direct P(Y) acquisition functionality is an important case that needs to be evaluated. The GPS receiver needs a precise time to perform direct P(Y) acquisition. The time accuracy, from a few nanoseconds to several milliseconds, has a strong impact on the GPS behavior. A special delay box applied to the pulse-per-second signal of the GPS simulator in accordance with PTTI message (that is, time figure of merit), enables such a simulated accuracy.
A standard IS 153-like interface was developed to display GPS data on a convenient GUI in order to have a common software to visualize output data from the GPS receivers. The user can also visualize some specific data from GPS ICDs concerning integrity, alarms, and cryptography.
All receiver output data are recorded for later analysis.
Monte Carlo Trials
The bench enables sequentially and automatically chaining scenarios (up to 10 000 test sequences) to perform statistics on acquisition times. Indeed, it is primarily used for the characterization of TTFF. GPS signal acquisition is dependent on many different parameters, as described in Figure 12. To properly characterize receiver acquisition times requires a large number of tests. The comparison with GPS Receiver Applications Module requirements can be easily performed.
One Monte Carlo trial consists of a repetition of unitary test: powering the receiver, then sending to the GPS receiver random errors of position, speed, time, levels of jamming, and finally stopping the test sequence on trigger. At the end of Monte Carlo trials, statistical computing enables accurate analysis and expertises.
The random selections are optimized to reduce the number of cases. The bench can replay a particular case: as the seeds are deterministic, a special case of Monte Carlo method can be selected and replayed.
Real-Time INS/GPS Data Fusion
The information delivered by INS and GPS are processed by a Kalman filter. The INS trajectory is provided by the simulator or by an external file.
Two types of coupling are considered: loose coupling with position and velocity information, and tight coupling with pseudoranges and delta ranges to estimate errors. In both cases, the GPS receiver receives aiding from either the simulated INS or the optimal navigation (Kalman filter output).
The purpose of the Kalman filter is to estimate the navigation errors (position, velocity, and attitudes) and sensor errors (INS, GPS).
The filter design is original because it is independent from the receiver under test and from the type of application (hardiness privileged with reference to jamming). It is also able to estimate the time offset between position and velocity measurement on any GPS receiver under test.
SINERGHYS combines several resources into a single test bench. A complex mode can simulate an operational implementation with different interfaces and by chaining test sequences: receiver initialization, management of the switching of antenna patterns during a simulation, masking of GPS signals, management of jamming, INS/GPS data fusion, and so on. In this mode, missions can be replayed in a realistic environment. This bench is a complementary resource for flight trials and digital models because it can characterize the initialization phases with a good control of initial conditions. SINERGHYS enables users to know, as precisely as possible, the capabilities and limitations of a specific global navigation chain.
SINERGHYS was developed by Bertin Technologies and specified by the French Ministry of Defense (MoD)DGA Information Superiority. It drives a Spirent GPS/Galileo simulator, Agilent 4431B and MXG generators, and software programs such as Analytical Graphics, Inc. (AGI) Satellite Tool Kit and Ergospace 3D scenes. The embedded jammer was developed by Ineo Defense in 2010 to MoD-DGA specifications.
Stéphane Gallot works at the French MoD (DGA Information Superiority) as a radionavigation expert. His particular interest is the integration of military GPS receivers including SAASM modules within French platforms.
Pascal Dutot is an architect engineer at the French MoD (DGA Information Superiority). His main activity is to optimize and control GPS integration in the global navigation chain.
Christophe Sajous works at the French MoD (DGA Information Superiority) as a radionavigation expert. He is also responsible for the “navigation per satellites” laboratory within the radionavigation department.