ION GNSS 2012
Session B1: GNSS Simulation, Testing & Timing Applications 1

Title: Expanding GNSS Testing with Multiple Synchronized Signal Recorders
Author(s): I. Ilie and S. Hamel, Averna Technologies, Canada
Date/Time: Wednesday, September 19, 2012, 8:35 a.m.
Room: 103/104 (NCC)

Record and playback systems bring valuable real-world test conditions into the lab because simulation is not enough to fully test some applications: they often require real-world signals and conditions to complete the test and validation processes. Thus, record and playback systems are recognized as precise and necessary tools during much GNSS receiver testing and validation, and they can also significantly reduce the amount of time and resources required for those phases.
Given the complexity of the GNSS spectrum, recorders used in the field should handle wideband signals as well as multiple carriers. For example, the GNSS spectrum covers the frequency band from 1164 MHz to 1300 MHz in the lower L band and from 1559 MHz to 1610 MHz in the upper L band. Inside this frequency band, a multitude of signals belong to different satellite constellations such as GPS, GLONASS, Galileo, and Compass.

To be able to efficiently record and play back all these signals, one needs an extremely wideband recorder or a multi-channel record and playback system with tight channel synchronization. As well, to synchronously record the GPS L1 and L2 bands, one needs two recording channels (since GPS L1 and L2 frequencies are separated by more than 300 MHz). In the same way, to cover the GPS L1, L2 and L5 bands, three channels are required. In some applications, the receiver is connected to multiple and/or multi-element antennas to take advantage of antenna selectivity or a multiple input approach (e.g., rake-like receivers). This approach requires even more recording channels to process the signal in a synchronous and coherent way.

To solve this synchronization and coherency challenge, Averna has developed a new software/hardware architecture that allows control and tight synchronization between multiple recording channels and multiple recording systems under the 1 nanosecond (ns) level. When Averna´s RP-5300 two-channel recorder is used, the architecture allows multiple RP-5300 chassis interconnections to form 4, 6, 8 or even more recording channels to operate in a synchronous and coherent way. Another feature of the architecture is the ability to control and synchronize multiple recording systems that are geographically separated. This is useful for Differential GNSS applications where the distance between the base and rover can reach tens of kilometers or even more. In fact, there is no limit to the distance that may separate multiple recorder systems, since the control is done over the Internet and all systems are synchronized with the unique GPS time reference.

In the first part of the proposed paper, the architecture is presented and the control and synchronization approaches are described. In the second part, two real use cases are presented: 1) wired synchronization and 2) wireless, geographically separated synchronization.

For the first use case, two RP-5300 two-channel recorder systems are wired to form a four-channel tightly synchronized recorder. Each recording channel is 50 MHz wide and allows multiple constellation (e.g., GPS + GLONASS) recording into the single channel. For validation purposes, all channels are connected with the same GNSS antenna via an active high-isolation splitter. A high-grade GNSS receiver is connected with the same GNSS antenna and the raw data is collected for comparison between the live and played-back signals.

The virtual four-channel recorder is configured to process GPS L1 + GLONASS G1 on channels 1 and 3, and GPS L2 + GLONASS G2 on channels 2 and 4. Once recorded, the data is played back with the four-channel player and the raw GNSS data is logged using the same high-grade reference GNSS receiver. In playback, to demonstrate the synchronization and coherency, the GPS L1 and L2 signals are recombined: GPS L1 from channel 1 with GPS L2 from channel 2, GPS L1 from channel 1 with GPS L2 from channel 4 and GPS L1 from channel 3 with GPS L2 from channel 2. It is expected that GPS L1/L2 signals will keep the coherence as per live recording and the synchronization error will be less than 1 ns for any L1/L2 combination.

In the second use case, the same two RP-5300 two-channel recorder systems are wirelessly interconnected using the GPS time from two separate GPS receivers as reference. The coherency and synchronization between the two recorders is based on the GPS-disciplined reference clock, 1 PPS signals, and extracted UTC time. The raw data in playback is collected and compared with live recording and the inter-channel synchronization error is calculated. It is expected that the synchronization error is in the same range as the 1 PPS signal´s precision (i.e., 1 PPS is supplied by GPS receivers that are part of each recorder system).

In conclusion, the proposed paper details the next stage of Averna´s record and playback architecture development, which allows the control of multiple recording systems and channels. It shows that with the new architecture, the multiple (N) RP-5300 two-channel recorder systems can be interconnected and tightly synchronized to form a virtual N x 2-channel recorder. The synchronization error between any two recorder channels is expected to be less than 1 ns. In addition, this paper shows that multiple recorder systems can be synchronized wirelessly using GPS as reference time with a precision of tens of nanoseconds, regardless of the distance between the systems.

Return to Session B1 Next Abstract