BLIND

The use of new signals for localisation and navigation is a current trend in research, e.g. to improve the accuracy and availability of localisation in urban environments. To do this, the receiver must know the signal structure (so-called PRN codes and modulation) of the satellite signals. This information is not available for non-GNSS satellite missions and must be estimated using a highly sensitive antenna.

The experts at the DIGITAL Institute therefore developed a concept demonstrator (CD) for a 40-element antenna system based on low-cost software-defined radio (SDR) platforms and commercially available components. Compared to a parabolic antenna, this approach has the major advantage that an omnidirectional antenna with high gain is able to track all satellites in the field of view.

The so-called antenna platform consists of an array of 40 identical off-the-shelf low-cost GNSS antennas connected to 20 dual-channel low-cost SDRs and 20 low-cost PCs for data storage. All loosely synchronised data is collected in a single database and synchronised and processed together with a software tool specifically designed for the recovery of unknown code chip sequences and running on a high-end PC with a 32-core CPU. The system was designed so that the chip error rate for the encrypted GNSS services of GPS, BeiDou and thus also for Galileo is sufficiently low - provided that the satellites are above a certain minimum elevation. A tilting mechanism for the antenna array was implemented to increase the achievable gain for low-altitude satellites. Finally, an existing GNSS software receiver was upgraded to implement the hybrid position, velocity and time solution by combining open signals with the blind tracked signals of the antenna array. For this purpose, a "receiver platform" was built consisting of four commercially available, low-cost antennas connected to two low-cost dual-channel SDRs and a commercially available GNSS receiver.

Algorithms and software were first verified with signals from a GNSS simulator. Numerous experiments were then carried out to test the performance with the signals in space - in particular the GPS M-code. The entire system was verified by cross-comparison with signals captured by a 2.4 metre steerable parabolic antenna. This showed that the tracking results finally obtained (C/N0, code/carrier pseudo-range, etc.) were in line with expectations. Finally, encrypted M-code signals (using the blindly estimated chip sequences of our system) were used as potential unknown L-band signals for localisation.

This array enables a variety of R&D activities, such as localisation with signals of opportunity, signal quality monitoring, direction finding, beamforming, antennas with controlled receive pattern. Possible future LEO navigation satellite systems transmitting in the L1 band can also be operated with this system. To use other frequency bands, only the antennas and LNAs need to be exchanged. The results indicate that reliable chip estimation can also be carried out for these other satellite systems and that these signals can be recorded and tracked, e.g. for a better understanding of higher-order BOC modulation methods.